Survey of Corticioid Fungi in North American Pinaceous Forests Reveals Hyperdiversity, Underpopulated Sequence Databases, and Sp

Mycologia

ISSN: 0027-5514 (Print) 1557-2536 (Online) Journal homepage: http://www.tandfonline.com/loi/umyc20

Survey of corticioid fungi in North American pinaceous forests reveals hyperdiversity, underpopulated sequence databases, and species that are potentially ectomycorrhizal

Lisa M. Rosenthal, Karl-Henrik Larsson, Sara Branco, Judy A. Chung, Sydney I. Glassman, Hui-Ling Liao, Kabir G. Peay, Dylan P. Smith, Jennifer M. Talbot, John W. Taylor, Else C. Vellinga, Rytas Vilgalys & Thomas D. Bruns

To cite this article: Lisa M. Rosenthal, Karl-Henrik Larsson, Sara Branco, Judy A. Chung, Sydney I. Glassman, Hui-Ling Liao, Kabir G. Peay, Dylan P. Smith, Jennifer M. Talbot, John W. Taylor, Else C. Vellinga, Rytas Vilgalys & Thomas D. Bruns (2017) Survey of corticioid fungi in North American pinaceous forests reveals hyperdiversity, underpopulated sequence databases, and species that are potentially ectomycorrhizal, Mycologia, 109:1, 115-127, DOI: 10.1080/00275514.2017.1281677 To link to this article: http://dx.doi.org/10.1080/00275514.2017.1281677

Accepted author version posted online: 19 Submit your article to this journal Jan 2017. Published online: 19 Jan 2017.

Article views: 327 View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=umyc20

Download by: [University of Wisconsin - Madison] Date: 13 October 2017, At: 20:06 MYCOLOGIA 2017, VOL. 109, NO. 1, 115–127 http://dx.doi.org/10.1080/00275514.2017.1281677

Survey of corticioid fungi in North American pinaceous forests reveals hyperdiversity, underpopulated sequence databases, and species that are potentially ectomycorrhizal Lisa M. Rosenthala, Karl-Henrik Larssonb, Sara Brancoc, Judy A. Chungd, Sydney I. Glassmand, Hui-Ling Liaoe, Kabir G. Peayf, Dylan P. Smithg, Jennifer M. Talboth, John W. Taylorg, Else C. Vellingag, Rytas Vilgalyse, and Thomas D. Bruns g aDepartment of Plant Pathology, University of California Davis, Davis, California 95818; bNatural History Museum, University of Oslo, Blindern, 0318 Oslo, Norway; cEcologie, Systematique et Evolution, Université Paris Sud, CNRS, AgroParisTech, Université Paris-Saclay, 91405 Orsay, France; dDepartment of Environmental Science Policy and Management, University of California Berkeley, Berkeley, California 94720; eDepartment of Biology, Duke University, Durham, North Carolina 27708; fDepartment of Biology, Stanford University, Stanford, California 94305; gDepartment of Plant and Microbial Biology, University of California Berkeley, Berkeley, California 94720; hDepartment of Biology, Boston University, Boston, Massachusetts 02215

ABSTRACT ARTICLE HISTORY The corticioid fungi are commonly encountered, highly diverse, ecologically important, and Receved 31 May 2016 understudied. We collected specimens in 60 pine and spruce forests across North America to Accepted 8 August 2016 survey corticioid fungal frequency and distribution and to compile an internal transcribed spacer KEYWORDS (ITS) database for the group. Sanger sequences from the ITS region of vouchered specimens were BLAST; corticioid fungi; compared with sequences on GenBank and UNITE, and with high-throughput sequence data from ectomycorrhizal; Picea; soil and roots taken at the same sites. Out of 425 high-quality Sanger sequences from vouchered Pinus; sequence survey specimens, we recovered 223 distinct operational taxonomic units (OTUs), the majority of which could not be assigned to species by matching to the BLAST database. Corticioid fungi were found to be hyperdiverse, as supported by the observations that nearly two-thirds of our OTUs were represented by single collections and species estimator curves showed steep slopes with no plateaus. We estimate that 14.8–24.7% of our voucher-based OTUs are likely to be ectomycorrhizal (EM). Corticioid fungi recovered from the soil formed a different community assemblage, with EM taxa accounting for 40.5–58.6% of OTUs. We compared basidioma sequences with EM root tips from our data, GenBank, or UNITE, and with this approach, we reiterate existing speculations that Trechispora stellulata is EM. We found that corticioid fungi have a significant distance-decay pattern, adding to the literature supporting fungi as having geographically structured communities. This study provides a first view of the diversity of this important group across North American pine forests, but much of the biology and taxonomy of these diverse, important, and widespread fungi remains unknown.

INTRODUCTION forming fungi are ectomycorrhizal (hereafter EM). In contrast, corticioid fungi were traditionally thought to be Despite major advances in fungal ecology due to the advent saprotrophic due to their association with dead plant mate-

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 of molecular techniques (Peay et al. 2008), very little pro- rial and consequently were ignored as potential EM associ- gress has been made in understanding the basic autecology ates (Larsen 1968, 1974). Although some early work of corticioid fungi, a nonmonophyletic group of showed that particular corticioid fungi were EM (Zak and Basidiomycota having a flat, crust-like morphology. Their Larsen 1978), it was a surprise when molecular ecology inconspicuous basidiomata and the limited number of peo- studies revealed them to be major components of EM pletrainedtoidentifythesefungiarethemajorreasonsfor communities (Kõljalg et al. 2000). In particular the the lack of studies. This is unfortunate because corticioid Atheliales, a group entirely composed of corticioid species, fungi are widespread and are known to fill multiple guilds are now widely accepted as dominant EM symbionts in as pathogens, saprotrophs, and ectomycorrhizal symbionts. many forest ecosystems. Erland (1995) found that There is a wealth of information supporting the view Tylospora spp. were the most common EM fungi in a that mushrooms, false truffles,truffles,andvariouscup-

CONTACT Thomas D. Bruns [email protected] Dept. Plant & Microbial Biology, 111 Koshalnd Hall, University of California, Berkeley, 94720-3102 USA. Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/umyc. Supplemental data for this article can be accessed on the publisher’s Web site. © 2017 The Mycological Society of America 116 L. M. ROSENTHAL ET AL.: NORTH AMERICAN CORTICIOID SURVEY

Swedish spruce forest, and a survey in Czech Republic that many of these unmatched sequences may be spruce forests showed that they colonized nearly 70% of corticioid fungi, and this idea motivated our study. the spruce root tips in disturbed sites (Peter et al. 2008). In To contribute to the knowledge of EM corticioid a North Carolina pine forest with elevated CO2 levels, fungi in North American pinaceous forests, we collected Tylospora spp. were also the most commonly recorded and sequenced specimens from 60 Pinus- and Picea- EM taxa, constituting over 30% of those recovered from dominated forests across the continent. Soil, EM roots, soil ingrowth bags (Parrent and Vilgalys 2007). Species of and pine seedling bioassays from a subset of these sites Piloderma, another genus in the Atheliales, are common in were previously studied with next-generation sequen- EM community studies (Tedersoo et al. 2008;Lilleskov cing (Talbot et al. 2013, 2014; Glassman et al. 2015), et al. 2011; Glassman et al. 2015) and have recently been rendering a wealth of molecular data suitable for inves- shown to be the most transcriptionally active EM fungi in tigating the prevalence of these fungi. Our goals were the some pine forests (Liao et al. 2014). following: (i) to obtain an initial view of the distribution The current view of fungal EM and saprotrophic and diversity of corticioid fungi, particularly those in the communities is increasingly driven by developments Atheliales, in North American pinaceous forests; (ii) to in the field of molecular ecology (Peay et al. 2008; help fill gaps in the sequence database of corticioid fungi; Lindahl et al. 2013). Recently, next-generation DNA and (iii) to identify additional taxa that are likely to be sequence analyses of colonized roots and soil have EM symbionts. been used to determine the identity and frequency of EM fungal species because this approach allows MATERIALS AND METHODS for much greater sample sizes and geographic cover- age (Talbot et al. 2014;Tedersooetal.2014). Study sites and collection.— Corticioid basidiomata However, a high percentage of these environmental were collected from 60 pine and spruce forest sites sequences often fail to match identified fungal across 15 North American states and provinces sequences present in the public databases and yet during the 2012 and 2013 summer months (Fig. 1A). are well matched to other environmental sequences Sites were typically monospecific stands of different from previous studies. This means that environmen- species in the Pinaceae family that were selected in tal sequences likely represent fungi that are common order to study the community structure of EM fungi enough to be resampled, but no information on on a local and the continental scale (Talbot et al. 2014) their trophic status can be inferred. We speculated (Table 1). At each site, basidiomata were collected at Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017

Figure 1. A. Distribution map of sample sites across North American pine forests (n = 60 sites). B. At each site sampling followed a 40 × 40 m2 nested square design where every corner designated a collection point (n = 13 samples). MYCOLOGIA 117

Table 1. Locations, samples types, and sequence numbers. Locationa Dominant treeb Year No. of sites Collection No. Sequence no.c Singleton no./total OTUsd AKsr PICg 2012 2 43 21 12/15 BC Pc, Pp PICe, PICe/Al 2013 7 125 61 28/38 CA-1sr Pa, Pc, Pmon, Pp 2012 6 72 41 22/30 CA-2sr Pmur 2012 3 40 30 21/25 CA-3 Pp, Pmur., and others 2011–2013 23 18/20 CO Pc, Pf, Pp PICe/Al 2013 7 68 25 19/22 CT Ps 2012 2 23 15 15/15 FL Pt, Pel 2012 2 32 12 10/11 MIs Pb, Ps, PICg, PICm/Ab 2013 6 47 27 17/20 MNs Pb, Pb/Pr 2012 2 30 14 10/12 MSs Pt, Pec 2012 2 27 13 10/11 MT Pa, Pc, PICe/Al 2013 5 73 44 22/31 NCs,r Pt 2012 2 38 23 18/19 ON Pb, Ps, PICg, PICm/Ab 2013 4 49 27 13/18 OR Pc, Pp 2012 4 45 32 16/23 VA Ps 2013 2 18 10 8/9 WY Pc, Pp, PICe 2013 4 27 9 9/9 Total 60 730 425 140/223 aAK: Alaska, USA; BC: British Colombia; CA-1: California, USA, Sierra Nevada plots; CA-2: California, USA, Point Reyes National Seashore plots; CA-3: California, USA, outside plots, Point Reyes, Yosemite, and Santa Cruz area; CO: Colorado, USA; CT: Connecticut, USA; FL: Florida, USA; MI: northern Michigan, USA; MN: Minnesota, USA; MS: Mississippi, USA; MT: Montana; NC: North Carolina, USA; ON: Ontaria, Canada; OR: Oregon, USA; VA: Virginia, USA; WY: Wyoming, USA. bPa: Pinus albicaulis; Pb: P. banksiana; Pc: P. contorta; Pec: P. echinata; Pel: P. elliottii; Pf: P. flexilis; Pmur: P. muricata; Pmon: P. monticola; Pp: P. ponderosa; Pr: P. resinosa; Pt: P. taeda; PS: P. strobus; PICe: Picea engelmannii; PICg: Picea glauca; PICm: Picea mariana; Al: Abies lasiocarpa; Ab: A. balsamifera. cNumber of sequences derived from vouchered collections. Next-generation sequences from soil, roots, or bioassays are not listed. dNumber of s derived from sequenced collections that are singletons (unique)/total s derived from collections. Superscript s and r indicate locations having sites with spore bioassay and EM roots sequenced, respectively. Sites sampled in 2012 have associated 454 soil sequences, and those from 2013 have Illumina sequences. Soil sequences were not collected from CA-3 sites.

13 points within a 40 × 40 m2 nested square (Fig. 1B). uncontaminated sequences. To circumvent this problem, If no basidioma was present at the sample point, the we picked very small tissue samples from each specimen search was expanded a few meters until the nearest under a dissecting microscope using sterile insect pins and one was found, and if none was still found within a placed each into 10 µL of extraction solution. After heating few meters, the sample point was skipped. When to 80 C for 15 min, 30 µL of dilution solution was added. fruiting was abundant, two collections per sampling Typically, 1 µL of final diluted extract was used in a 25-µL point were made and additional samples were reaction to amplify the internal transcribed spacer (ITS) collected from the general area. In efforts to sample region using the standard fungal primers ITS-1F and ITS-4 EM corticoid fungi, we tried to target specimens in the (White et al. 1990;GardesandBruns1993). The Atheliales, which usually have a sparse, yet smooth and Basidiomycete-specific primer ITS-4B (Gardes and Bruns continuous hymenium. Collectors examined well- 1993) was occasionally used instead of ITS-4 to exclude decayed wood and litter, selecting the first light- Ascomycete contaminants. Polymerase chain reaction colored corticioid basidioma with a cobweb-like (PCR) products yielding a single band during gel appearance. When fungi with a light-colored and electrophoresis were purified using ExoSAP-IT (Sigma thin morphology could not be found, we also Aldrich) and cycle-sequenced in both directions with the sampled corticioid fungi of various morphologies and BigDye Terminator v3.1 kit (Applied Biosystems, Foster resupinate polypores. Finally, we augmented our City, California, USA). Sequences were read with the ABI Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 collection with an additional 23 samples that were 3730 DNA analyzer (Applied Biosystems). Bidirectional collected along the California coast and in the sequences were edited using Sequencher (GeneCodes, Yosemite region of the Sierra Nevada mountains. Ann Arbor, Michigan, USA) and Geneious Pro v5.4.6 Each specimen was photographed, saved for DNA (BioMatters, Auckland, New Zealand) and aligned to extraction, and dried as a voucher to be deposited in form contiguous sequences. These were deposited in the University of California (UC) Herbarium GenBank under accession numbers KP814141–KP814565. (UC2022802–UC2023243).

Taxonomic identification of basidiomata.— DNA extraction, PCR, and sequencing of Basidioma sequences including ITS1-5.8s-ITS2 regions basidiomata.— DNA was extracted using the Sigma were grouped into operational taxonomic units (OTUs) Extract-N-Amp kit (Sigma Aldrich, St. Louis, Missouri, using the UCLUST method (Edgar 2010)inQIIME USA). The basidiomata were often covered in other (Caporaso et al. 2010). We compared similarity cutoffs of material that can hinder the ability to obtain high-quality, 95–99% and found a linear relationship between the 118 L. M. ROSENTHAL ET AL.: NORTH AMERICAN CORTICIOID SURVEY

number of resulting OTUs and percent sequence identity included all OTUs, treated them as if they were drawn in the 95–97% range, followed by an upward inflection in from a common pool, and produced a rarefaction curve the 98–99% range. For that reason, we adopted the 97% and a Chao2 species estimator curve (Chao 1987)using cutoff as a conservative estimate of species. To reduce the program EstimateS (Colwell 2013). artifactual OTU inflation caused by the software, sets of OTUs that BLASTed to the same sequences were manually aligned and combined if the sequence similarity was Determination of EM Fungi. —Given that the greater than 97%. taxonomic assignments were often imprecise and our Representative sequences were compared against the knowledge of the nutritional modes of many taxa is still UNITE and NCBI GenBank databases using the in a state of flux (Rinaldi et al. 2008), we adopted two massBLASTer search option on the UNITE interface approaches to estimate the proportion of EM corticioid (Kõljalg et al. 2013). We chose reference sequences to fungi in our data set. Conservatively, OTUs were infer taxonomic identification by placing more confi- classified as EM only if they BLAST-matched to dence in UNITE voucher-based sequences and those established EM taxa listed in Tedersoo et al. (2010). derived from in-depth systematics studies. Less conservatively, we classified OTUs as possibly EM A consensus was reached for each OTU identity, depen- if they matched genera that include EM species, even if dent upon the sequence similarity and taxonomic annota- congeners have other nutritional modes (e.g., tion of the BLAST match. Generally, we adopted the same Sistotrema, Sebacina). Additionally, we classified any species name of matches with a sequence similarity greater OTU as potentially EM if it matched sequences from than 97% and the same genus name for matches that were EM root tips in GenBank or EM root tips from our in- more than 95% similar. Since ITS variation fluctuates house EM spore bank bioassay (Glassman et al. 2015) across groups and many of our OTUs matched unresolved and root database with at least a 95% similarity. species complexes, we assessed the identity of OTUs on an individual basis and did not rely solely on a hard cutoff. Individual naming decisions are explained more comple- tely in the comments following Supplementary Table 1. In Connecting aboveground and belowground — some cases when sequence-based identification failed, we sequences. We connected our specimen-based could assign names by morphologically identifying speci- sequences to ITS sequences from soil cores, spore mens using Bernicchia and Gorjón (2010). Positive identi- bioassays, and root samples for three reasons: (i) to fications were limited by the difficult morphological understand the spatial patterns of corticoid fungi, (ii) taxonomy of these fungi, and the absence of distinctive to provide insight into the belowground activity of cystidia, hyphae, basidia, and spores in some specimens. corticoid fungi, and (iii) to compare the taxonomic Corticioid taxonomy is very much a work in progress, diversity of corticioid fungi aboveground and resulting in many nonmonophyletic groups. In light of belowground. Soil was collected from the O and A this, we assigned tentative names to some OTUs, adding horizons from the same collection sites, DNA was sensu lato when they matched to a species that fell outside extracted, and the ITS region of rDNA was amplified. of the sensu stricto clade, or when it was morphologically DNA from soil samples collected at sites visited in 2012 identified to a group but the sequence lacked a close match. (n = 25 sites) was sequenced on the 454 platform as previously described and reported (Talbot et al. 2013,

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 All names were updated to the current nomenclature in IndexFungorum (http://www.indexfungorum.org/). OTUs 2014), and that from sites visited in 2013 (n = 35) was that matched well to non-corticioid Basidiomycota or processed via Illumina sequencing (unpublished data). Ascomycota were assumed to be contaminants or sterile Implemented on the UPARSE pipeline (Edgar 2013), mycelium and were excluded. these sequences were quality filtered, denoised, trimmed, and paired as in Smith and Peay (2014). Soil collected from 19 sites for EM spore bank bioassays, as Analysis of taxonomic distribution and alpha reported by Glassman et al. (2015), and EM roots of diversity.— A ranked abundance curve was constructed resident trees from 9 sites were collected and DNA was for samples across all genera to understand the taxonomic extracted, the ITS region of rDNA was amplified, and distribution of our basidioma collection. Those that were amplicons were sequenced via 454 pyrosequencing as best identified to a group higher than genus were an unpublished part of the Talbot et al. (2014) study. considered unidentified. The curve also served to Details regarding which sites are associated with the directly compare the diversity of aboveground and soil cores, EM spore bank bioassays, and root samples belowground corticioids. To assess species richness, we are provided in Table 1. MYCOLOGIA 119

Before merging the 454 and Illumina data to com- Finally, the soil sequence data set was used to com- plete the soil sequence data set, ITS1 was extracted pare the taxonomic diversity between aboveground and from the 454 sequences so that sequences from both belowground communities. In deciding a common platforms were of comparable length. OTUs were then metric between soil and basidioma sequences, we picked with the UPARSE method in USEARCH (Edgar chose presence/absence in soil core samples instead of 2010), using the standard settings of a 97% sequence read abundance, because the read abundance has a very similarity, de novo and reference-based chimera check- weak relationship with quantity across different species ing, and removal of singletons. OTUs were taxonomi- (Amend et al. 2010; Nguyen et al. 2014). cally assigned with the BLAST method in QIIME, utilizing reference data sets and id-to-taxonomy maps RESULTS from fungal database UNITE (Kõljalg et al. 2005; http:// unite.ut.ee/index.php). We generated a list of known Taxonomic identification of ITS sequences.—In corticioid genera and family- or ordinal-level corti- total, we made 757 collections of corticoid cioid-exclusive taxonomic groups from Bernicchia and basidiomata. The number of collections per 40 × Gordon (2010) and from two large-scale phylogenetic 40 m2 plotrangedfrom3to32withameanof studies of corticioid fungi by Larsson et al. (2004) and 14.6 per plot. From these 757 vouchers, we obtained Binder et al. (2005). Soil OTUs that matched names in 498 sequences, which represents a success rate of this list were identified and extracted. In order to iden- 66% for sequence acquisition. A preliminary tify any soil OTUs that were not previously included BLAST analysis identified 63 sequences that due to omissions in the public databases, we repicked matched non-corticioid taxa. These were dropped OTUs from the soil sequences using the ITS1 extracted from subsequent analyses, leaving 425 high-quality region of our vouchered corticioid OTUs as a reference bidirectional sequences. OTUs were picked using a database, utilizing QIIME and a 97% identity cutoff. 97% similarity threshold, followed by individually Augmented with any additional OTUs matching above- aligning and inspecting putative OTUs that ground corticioid fungi, this soil sequence data set BLASTed to the same sequence. The latter resulted represented our belowground corticoid sequences. in merging 38 erroneously split groups, resulting in The comprehensively sampled set of soil sequences a total of 223 OTUs. Representative sequences of afforded us the ability to analyze the relationship between each OTU were compared against UNITE and community dissimilarity and physical distance of corti- GenBank databases for taxonomic assignments cioid fungi. Sample sequence depth ranged tremendously; (Supplementary Table 1). thus, we rarefied the samples to the 10th and 15th percen- A total of 61.9% of OTUs matched to a species-anno- tile sequence depths (1145 and 1663 sequences, respec- tated sequence at 97% identity or higher. Of the OTUs with tively). Utilizing the beta_diversity.py command in poor matches, we morphologically identified 36 OTUs to at QIIME, OTU tables were rarefied to both levels and least the generic level, leaving only 8 OTUs identified to three beta-diversity indices (the presence/absence-based either class or order. Among this set of morphologically Sorensen-Dice and Jaccard indices and the abundance- identified specimens, Tubulicrinis was the dominant genus based Bray-Curtis index) were produced as matrices of with 9 OTUs, followed by Botryobasidium with 5 OTUs. pairwise comparisons between each sample site for both Many of these OTUs that we morphologically identified as

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 rarefication levels. Mantel tests, from the vegan package Tubululicrinis or Botryobasidium had BLAST matches to (Oksanen et al. 2012)inRversion3.0.1(RDevelopment the correct genus, but with a similarity that was often below Core Team 2013), compared these metrics with the physi- 90%. Although we arrived at a name for the majority of cal distance matrix, produced from the geosphere package samples, identification by DNA barcoding and morphology (Hijmans et al. 2012). Mantel correlograms from vegan is imperfect and we indicate those that should be inter- were used to detect distance-decay pattern and to provide pretedcautiously,eitherbycf.,aff.,ors.l.Weexaminedthe insight into the scale at which species turnover occurs. taxonomic distribution of our samples by genus To identify corticioid taxa that are potentially EM, (Fig. 2). The 68 genera recovered spanned all 12 we looked for voucher sequences that matched major Agaricomycete (previously considered sequences from colonized root tips in the plots or Homobasidiomycete) clades established by Binder et al. colonized root tips from EM spore bioassays. We (2005). The ranked abundance curve shows a handful of employed the same reference-based OTU picking pro- dominant genera, followed by a long tail of rare taxa cedure described above to connect ITS sequences from (Fig. 2). The nine most dominant genera accounted for EM spore bioassays and EM root samples to our speci- over 50% of identified samples. In order of decreasing men-based sequences. abundance, these were Tubulicrinis, Piloderma, Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 aaaeidctdb n * respectively. **, and * by indicated are taxa display 2. Figure 120 eaiepooto fsmlsa tce as oddtx eefudecuieya aiimt n Mo oetal EM potentially or EM and basidiomata as exclusively found were taxa Bolded bars. stacked as samples of proportion relative akdaudnecreo eeao h aiim n olcmuiis aiim bak n ol(ry communities (gray) soil and (black) Basidioma communities. soil and basidioma the of genera of curve abundance Ranked .M OETA TA. OT MRCNCRIII SURVEY CORTICIOID AMERICAN NORTH AL.: ET ROSENTHAL M. L. Taxa Unidentified corticioid Marchandiobasidium Acanthophysellum Confertobasidium Botryohypochnus Pseudotomentella* Marchandiomyces Leptosporomyces Gloeocystidiellum Pseudomerulius Ceratobasidium** Cylindrobasidium Fibulorhizoctonia Chaetodermella Leucogyrophana Aphanobasidium Scytinostromella Sistotremastrum Byssocorticium* Tomentellopsis* Basidiodendron Botryobasidium Phanerochaete Amyloxenasma Thanatephorus Hymenochaete Diplomitoporus Hydnomerulius Ceraceomyces Radulomyces Cabalodontia Scytinostroma Crustomyces Hypochnicium Wrightoporia Physisporinus Limonomyces Steccherinum Brevicellicium Fibulomyces Peniophorella Xenasmatella Grammothele Dacryobolus Heterochaete Asterostroma Trechispora** Haplotrichum Serendipita** Hastodontia Hyphoderma Ramaricium Amphinema* Perenniporia Hyphodontia Meruliopsis Sistotrema** Truncospora Dendrothele Scopuloides Punctularia Thelephora* Bjerkandera Gloeodontia Rigidoporus Tomentella* Rhizoctonia Amaurodon Coniophora Junghuhnia Peniophora Flaviporus Fuscoporia Veluticeps Sebacina** Tubulicrinis Piloder Resinicium Kneiffiella Vuilleminia Antrodiella Tylospora* Dentipellis Fibroporia Gloiothele Mycoacia Fibricium Lyomyces Ceriporia Phlebiella Corticium Phellinus Xylodon Antrodia Boidinia Serpula Phlebia Vararia Kavinia Athelia Jaapia Sidera Exidia Irpex ma* .000 .001 .002 0.30 0.25 0.20 0.15 0.10 0.05 0.00 Above−ground andbelow−ground communities Proportion ofsamples Proportion Soil Basidiomata MYCOLOGIA 121 Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 Figure 3. Examples of collected and sequenced basidiomata that resemble the athelioid morphology. (A) OTU 52 Kneiffiella sp.; (B) OTU 25 Ceraceomyces cf. serpens; (C) OTU 200 Atheliaceae; (D) OTU 253 Trechispora sp. 3; (E) OTU 27 Sistotrema sp.; (F) OTU 148 Leptosporomyces aff. raunkiaeri; (G) OTU 87 Hyphodontia floccosa; (H) OTU 153 Botryobasidium sp. 1. Dissecting scope views ~30× shown in E, G, and H.

Amphinema, Trechispora, Botryobasidium, Hyphodontia, OTUs and 33.4% of samples belonged to other genera that Leptosporomyces, Peniophorella,andAthelia. also have a light-colored and thin hymenium (Fig. 3)but The Atheliales, in accordance with Larsson et al. (2004) are not in the Atheliales. This set included species in and Binder et al. (2010), represented 17.0% of OTUs and Botryobasidium (11 OTUs, 23 samples), Ceraceomyces (7, 23.8% of samples and were composed of the following: 8), Hyphoderma (7, 10), Hyphodontia (11, 22), Sistotrema Amphinema (6 OTUs, 26 samples), Athelia (7, 14), (5, 5), Sistotremastrum (2, 9), Trechispora (12, 26), and Piloderma (11, 29), Leptosporomyces (8, 22), Tylospora (4, Tubulicrinus (18, 39). In total, 49.7% of OTUs and 57.2% 6), and unidentified Atheliales (2, 4). A total of 32.7% of of the collections had the targeted morphology. 122 L. M. ROSENTHAL ET AL.: NORTH AMERICAN CORTICIOID SURVEY

Fungi from soil. —The soil sequences contained a curve of basidioma collections was nonasymptotic, diverse corticioid assemblage that partially and the Chao 2 species estimator curve did not overlapped with aboveground fungi and was plateau (Fig. 4). Although species accumulation enriched in EM fungi (Fig. 2). In total, we found curves were not shown for individual sampling 938 OTUs representing 3,423,152 corticioid regions, they too exhibited similar steep patterns sequences in the soil cores. The overwhelming because all but two regions were comprised of over majority,96.8%,wereidentifiedviaUNITE 50% singletons (see Table 1). reference data sets and id-to-taxonomy maps, whereas 3.2% matched only our voucher collections. Ectomycorrhizal taxa. —Based on our conservative and Only 108 of the 223 (48.3%) vouchered OTUs were relaxed criteria, we estimated that 14.8–24.7% of also found in the soil. This limited overlap is vouchered OTUs and 17.9–26.8% of collections were surprising because there were more than four times EM. Of the 108 OTUs that were also recovered in the as many soil OTUs than voucher-based OTUs. Out soil, 25.9–34.3% belonged to EM genera. A larger of the 80 genera represented in the soil sequences, 35 proportion of the soil community was EM, constituting were found exclusively in the soil samples. The 40.5–58.6% of OTUs and 61.3–79.3% of samples. Several taxonomic distribution of corticioids in the soil was vouchered OTUs matched EM roots in GenBank, heavily dominated by a small number of EM genera, UNITE, our plots, or our seedling bioassays. Most of the with Piloderma, Amphinema, Tomentella, and identified taxa that matched root sequences were from Pseudotomentella, in ranked order of abundance, known ectomycorrhizal genera: Amphinema, Piloderma, present in over 50% of the soil samples. Conversely, Tylospora, Pseudotomentella, Thelephora, Tomentella, 23 genera were only found in the voucher collection Tomentellopsis, Sistotrema, and Sebacina. Trechispora (see bolded taxa, Fig. 2), and this set of taxa appears was also found to match sequences from the soil data set to be enriched in non-EM taxa. and EM roots (Supplementary Table 1), and as discussed below, there is reason to reaffirm previous suspicions that some species are EM (see Dunham et al. 2007). — Species richness. Our basidioma collection was Hypochnicium, Scytinostroma, and Tubulicrinis matched predominately composed of singletons, representing EM roots, but based on current knowledge of their 62.8% of all OTUs (140/223 OTUs), and only eight of ecology, we did not consider them potentially EM. the OTUs were collected at five or more sites (Supplementary Table 1). Due to the high — percentage of singletons, the species accumulation Distance-decay pattern. There was a significant effect of distance decay on our soil collection of corticioid fungi (Figure 5A). Among the set of three beta-diversity indices analyzed at two rarefaction levels, Mantel r ranged between 0.348 to 0.589 and P < 0.001 (Supplementary Table 2), indicating that a distance- decay pattern was present regardless of the parameters. The mantel correlogram, analyzed using the Jaccard

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 dissimilarity index at the lower rarefaction level, showed corticioid fungal assemblages to be significantly different after 1000 km (Fig. 5B, Supplementary Table 3).

DISCUSSION

Our continental survey of corticioid fungi added 425 new sequences representing 223 OTUs to the ITS database, and when these voucher sequences were combined with EM root sequences, our study enhances our understanding of the autoecology and distribution of corticioid fungi. For such an overlooked group of fungi whose morphol- Figure 4. Rarefaction curve with 95% CI (black) and Chao 2 ogy offers few diagnostic characters, the addition of 215 curve (gray) for entire voucher collection. new OTUs identified to at least genus level is significant MYCOLOGIA 123

Figure 5. Analysis of spatial autocorrelation. A. Mantel test of Jaccard community dissimilarity versus geographic distance (m). Points represent pairwise comparisons between sites. Points are semi-transparent in order to better display those that are overlapping. B. Mantel correlogram shows the scale of species turnover. All points were significant (P < 0.05). Those above the zero line are more similar and those below more dissimilar than expected from a random distribution.

because molecular identification is the common method supporting that they may be mycorrhizal. Hypochnicium through which these taxa are now encountered in ecolo- species have been found as root endophytes (Chlebicki gical studies. Furthermore, our results reinforce the view 2009), perhaps explaining why its sequences were that a large proportion of corticioid fungi are present in retrieved from EM roots. Previous studies have posited EM communities (Kõljalg et al. 2000). Scytinostroma and Tubulicrinis as EM based on root tip Among the EM corticioid fungi in our study, morphology (Rinaldi et al. 2008). De Román and De Trechispora stellulata emerged as potentially EM because Miguel (2005) morphotyped Quercus EM root tips, sequences from basidiomata matched those of EM root assigning one type to Scytinostroma, but there was no tips. The nutritional modes of Trechispora are still incon- apparent effort made to link the morphotype to a mole- clusive, but recent findings suggest that some species, cular fungal identity. EM status of Scytinostroma is not including T. stellulata, may be mycorrhizal. An unidenti- supported based on its phylogenetic position (Miller et al. fied ericoid mycorrhizal Agaricomycete was found to be 2006), but given the polyphyletic nature of EM fungi, this sister to Trechisporales (Vohník et al. 2012). Assuming conclusion is not entirely convincing. Kernaghan (2001) ericoid mycorrhizae are associated with this clade, a tentatively identified Tubulicrinis as EM due to its dis- switchover to EM habit could have evolved indepen- tinctive amyloid lyocystidia, but in our samples only 1 out dently, as this is a common pattern (Tedersoo et al. of 18 Tubulicrinis OTUs matched to any root sequence.

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 2010). Additionally, Dunham et al. (2007)foundT.cf. Our observations in conjunction with these previous stellulata forming EM mats in pinaceous forests. reports leave the EM status of these three taxa ambiguous. Tedersoo et al. (2010) disputed its EM status, claiming Although a substantial subset of corticioid basidio- that tissue was collected from mycelial mats rather than mata was found to be EM, roughly 3 times as many soil root tips. In the study by Dunham et al. (2007), root tips samples were found to be EM. In fact, 50% of soil were indeed collected from every EM mat and morpho- corticioid samples were composed of four known EM logically confirmed identical to rhizomorphs before DNA genera, Piloderma, Amphinema, Tomentella, and extraction. Here we were not sampling in mats, but only Pseudotomentella. On the other hand, out of the nine root tips and basidiomata. T. stellulata from our collection genera that make up half of the vouchered samples, only had a 99% sequence similarity to Pinaceae EM root tips Piloderma and Ampinema are known EM fungi. collected by Dunham et al. (2007) and Peay et al. (2010). The disconnect between the aboveground and below- OTUs of Hypochnicium geogenium, Scytinostroma sp., ground fungal communities has been described on multi- and Tubulicrinis cf. sororius also matched some root ple accounts (Gardes and Bruns 1996; Smith et al. 2011) samples, but in these cases, we find no convincing data and was not unexpected. Still the pattern is striking, with 124 L. M. ROSENTHAL ET AL.: NORTH AMERICAN CORTICIOID SURVEY

less than 50% of voucher-based OTUs shared in common voucher-based OTUs (25.9–34.3% vs. 15.7–23.3%), and with the soil and large differences in frequency of the taxa every EM or potentially EM genus found as a basi- that do overlap (Fig. 2). In our study, these differences are dioma was also recovered in the soil. This pattern likely due to both sampling methods and autecology of the indicates that the frequency in soil may be one addi- taxa involved. Taxa found in the soil but not in the fruiting tional way to crudely identify other species as poten- record are likely attributable in part to a sampling artifact. tially ectomycorrhizal. By visiting each plot only once to sample basidiomata, Studies have increasingly demonstrated that fungal which are known to be temporally sensitive (Halme and community composition is spatially autocorrelated Kotiaho 2012), we obtained an incomplete snapshot of the (Lilleskov et al. 2004; Bahram et al. 2013), and the entire aboveground diversity. In contrast, the below- corticioid guild again demonstrates this pattern. Soil ground diversity varies much less throughout the seasons fungal communities (Talbot et al. 2014) and EM spore (Smith et al. 2011), possibly due in part to soil samples bank communities (Glassman et al. 2015) sampled at detecting metabolically inactive fungi (Baldrian et al. the same 2012 sites were found to have high levels of 2012). A more important factor is that our selective sam- geographic endemism across North America, and pling scheme targeted fungi that looked similar to the strong distance-decay relationships were consistent largely EM Atheliales. Because corticioid fungal groups throughout the local meter scale to the continental can seldom be identified by macromorphology, especially scale. Likewise, our study showed that corticioid fungi in the field, we recovered a high proportion of specimens exhibited a significant distance-decay relationship at with gross similarity to the Atheliales that belonged to the continental scale. Given that our sequences were distant orders. Ultimately, our collection represented a partially derived from the same soil sequences from phylogenetically diverse set of corticoid fungi that were Talbot et al. (2014), we are not providing independent biased towards light-colored and thin morphologies (57% evidence of the pattern, but we are showing that corti- of samples). Dark-colored species are underrepresented, cioid fungi are another group concordant with it, and most notably excluding many EM Thelephorales and per- the spatial turnover in species that we observe likely haps Byssocorticium. Therefore, a cautious interpretation explains at least part of the high species richness seen in of comparisons of above- and belowground communities our survey (Qian et al. 2005). It is generally understood is advised. that both environmental heterogeneity and dispersal The reverse pattern—taxa found fruiting but not in limitations dictate the extent of distance-decay patterns the soil—cannot be explained by the same sampling (Verleyen et al. 2009; Hanson et al. 2012), but several biases and instead points toward a biological difference microbial studies conducted on the regional to the in the two sample types. This conjecture is supported global scale identify dispersal limitations as the primary by the fact that the soil samples comprise primarily EM cause (Whitaker et al. 2003; Green et al. 2004; Reche fungi, whereas the fruiting samples have more sapro- et al. 2005; Talbot et al. 2014). With our corticioid trophs. We acknowledge that the Thelephorales are sample we cannot separate environmental drivers underrepresented in the fruiting samples, therefore arti- from dispersal limitation, but given the mode of spore ficially reducing the proportion of EM samples. If we dispersal in these fungi, it would not be surprising if take this sampling bias into account by removing sam- dispersal limitation was the dominant factor causing ples in the Thelephorales from both the above- and the distance-decay relationship.

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 belowground collections, the proportion of EM samples Both rarefaction and Chao2 species estimator curves in the soil is still appreciably greater (14.9–24.1% vs. did not plateau, demonstrating an undersampled, hyper- 52.7–74.7%). This observation is consistent with the diverse community. These curves closely corresponded idea that many decomposers may be largely confined to those based on other hyperdiverse fungal systems, to their carbon resources and do not send mycelium such as morphotypes of tropical endophytes and pyro- into the soil. In fact, all of the 17 genera that were sequences of Quercus leaf phyllospheres (Arnold et al. found exclusively in the voucher collection are thought 2000; Jumpponen and Jones 2009). Although it is not to be wood decayers or other saprotrophs (Fig. 2, possible to directly compare our data with these previous bolded text). Conversely, mycorrhizal fungi explore studies due to sampling and methodological differences, the soil to scavenge for mineral nutrients and to locate it is worth noting the similarities. Jumpponen and Jones uncolonized root tips and for that reason are more (2009) drew parallels between their Quercus leaf results likely to be detected in the soil sequences. In corrobora- and the tropical endophytes from the Arnold et al. tion of this explanation, the OTUs that were shared (2000) study, arguing that the fungal phyllosphere com- between the soil and voucher datasets were composed munity of temperate Quercus leaves was also hyperdi- of a higher proportion of EM taxa than the entire set of verse based on an equally high proportion of singletons. MYCOLOGIA 125

In both of these studies, over half of the fungi were Arnold AE, Maynard Z, Gilbert GS, Coley PD, Kursar TA. encountered as singletons, which is similar to the 62% 2000. Are tropical fungal endophytes hyperdiverse? Ecology – singleton rate in our corticioid collections. Letters 3:267 274, doi:10.1046/j.1461-0248.2000.00159.x Bahram M, Kõljalg U, Courty PE, Diédhiou AG, Kjøller R, Initial comparisons against public sequence databases Põlme S, Ryberg M, Veldre V, Tedersoo L. 2013. The distance led to a majority of our collections insufficiently identi- decay of similarity in communities of ectomycorrhizal fungi fied, and only after the UNITE database was updated in different ecosystems and scales. Journal of Ecology during the last few months with annotated sequences 101:1335–1344, doi: 10.1111/1365-2745.12120 from Europe were we able to attain species names for a Baldrian P, Kolarik M, Stursova M, Kopecky J, Valaskova V, Vetrovsky T, Zifcakova L, Snajdr J, Ridl J, Vlcek C, significant portion. Although this was not unforeseen, Voriskova J. 2012. Active and total microbial communities especially for a neglected group such as corticioid fungi, in forest soil are largely different and highly stratified it is indicative of a major problem because sequences are during decomposition. ISME Journal 6:248–258. becoming the standard means for fungal species identifi- Bernicchia A, Gorjón SP. 2011. Corticiaceae s.l. Fungi cation. Although much can be learned about the patterns Europaei 12. Alassio, Italy: Edizioni Candusso. 1008 p. of fungal diversity without attaching species names, what Binder M, Hibbett DS, Larsson K, Larsson E, Langer E, Langer G. 2005. The phylogenetic distribution of resupi- is missed with anonymous sequences is the ability to nate forms across the major clades of mushroom-forming retrieve and organize information about the autecology fungi (Homobasidiomycetes). Systematics and Biodiversity of the individual species (Peay 2014). 3:113–157, doi: 10.1017/S1477200005001623 We picked up a significant diversity in our basi- Binder M, Larsson K-H, Matheny PB, Hibbett DS. 2010. dioma surveys that was not recovered in the soil sam- Amylocorticiales ord. nov. and Jaapiales ord. nov.: early diverging clades of Agaricomycetidae dominated by cor- ples, despite the enormous sequencing capabilities of ticioid forms. Mycologia 102:865–880, doi: 10.3852/09- next-generation sequencing techniques. We were able 288 to apply names to many of these samples and docu- Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, ment their occurrence across a large area of the con- Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich tinent. Perhaps more importantly, the deposited K, Gordon JI, Huttley GA, Kelley ST, Knights D, Jeremy E, sequences will allow others to objectively compare Ley RE, Lozupone CA, Mcdonald D, Muegge BD, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA. 2010. QIIME what we found, independent of the names we apply, allows analysis of high-throughput community sequencing and then retrieve and study the collections of interest. data. Nature Methods 7:335–336, doi: 10.1038/nmeth. We hope this survey paves the way for future studies on f.303.QIIME these diverse, important, and widespread fungi. Chao A. 1987. Estimating the population size for capture- recapture data with unequal catchability. Biometrics 43:783–791, doi: 10.2307/2531532 ACKNOWLEDGMENTS Chlebicki A. 2009. Some endophytes of Juncus trifidus from Tatra Mts. in Poland. Acta Mycologica 44:11–17. We thank the following people for helping us with site selection Colwell RK. 2013. EstimateS: statistical estimation of species ’ and local logistics: D. Lee Taylor, Peggy Moore, Alison Colwell, richness and shared species from samples. User s guide Jennie Haus, Ben Becker, Matt Smith, Jason Hoeksema, Erik and application. Available from: http://purl.oclc.org/ Lilleskov, Richard Wilson, Cathy Cripps, Don Bachman, estimates. Melanie Jones, Dan Durall, Steve Miller, Jeff Mitton, Noah De Román M, De Miguel AM. 2005. Primeros datos sobre la Fierer, Robert Marra, Anthony D’Amato, Chuck Kramer, reforestación de un área de carrascal quemado con plantas Joshua Kragthorpe, and Shawn Fraver, and we thank Karen de Quercus ilex subsp. ballota inoculadas con Tuber mel-

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 Nakasone for co-hosting the NAMA workshop on corticioids, anosporum. Publicaciones de Biología, Universidad de – which helped motivate this work. Funding was provided by Navarra, Serie Botánica 16:19 40. National Science Foundation Dimensions of Biodiversity pro- Dunham SM, Larsson K-H, Spatafora JW. 2007. Species gram (awards 1045658, 1046115, and 1046052). richness and community composition of mat-forming ectomycorrhizal fungi in old- and second-growth Douglas-fir forests of the HJ Andrews Experimental – ORCID Forest, Oregon, USA. Mycorrhiza 17:633 645, doi: 10.1007/s00572-007-0141-6 Thomas D. Bruns http://orcid.org/0000-0002-2943-8669 Edgar RC. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–2461, doi: 10.1093/bioinformatics/btq461 Edgar RC. 2013. UPARSE: highly accurate OTU sequences from LITERATURE CITED microbial amplicon reads. Nature Methods 10:996–998. Eriksson J, Ryvarden L. 1975. The Corticiaceae of North Amend AS, Seifert KA, Bruns TD. 2010. Quantifying micro- Europe. Vol. 3. Oslo, Norway: Fungiflora. p 288–546. bial communities with 454 pyrosequencing: does sequence Erland S. 1995. Abundance of Tylospora fibrillosa ectomycor- abundance count? Molecular Ecology 19:5555–5565. rhizas in a South Swedish spruce forest measured by RFLP 126 L. M. ROSENTHAL ET AL.: NORTH AMERICAN CORTICIOID SURVEY

analysis of the PCR-amplified rDNA ITS region. Larsson K-H, Larsson E, Kõljalg U. 2004. High phylogenetic Mycological Research 99:1425–1428, doi: 10.1016/S0953- diversity among corticioid homobasidiomycetes. 7562(09)80788-0 Mycological Research 108:983–1002, doi: 10.1017/ Gardes M, Bruns TD. 1993. ITS primers with enhanced S0953756204000851 specificity for basidiomycetes—application to the identifi- Liao H-L, Chen Y, Bruns TD, Peay KG, Taylor JW, Branco cation of mycorrhizae and rusts. Molecular Ecology 2:113– S, Talbot JM, Vilgalys R. 2014. Metatranscriptomic ana- 118, doi: 10.1111/j.1365-294X.1993.tb00005.x lysis of ectomycorrhizal roots reveals genes associated Glassman SI, Peay KG, Talbot JM, Smith DP, Chung JA, with Piloderma-Pinus symbiosis: improved methodolo- Taylor JW, Vilgalys R, Bruns TD. 2015. A continental gies for assessing gene expression in situ. view of pine-associated ectomycorrhizal fungal spore Environmental Microbiology 16:3730–3742, doi: banks: a quiescent functional guild with a strong biogeo- 10.1111/1462-2920.12619 graphic pattern. New Phytologist 205:1619–1631. Lilleskov EA, Bruns TD, Horton TR, Taylor D, Grogan P. Green JL, Holmes AJ, Westoby M, Oliver I, Briscoe D, 2004. Detection of forest stand-level spatial structure in Dangerfield M, Gillings M, Beattie AJ. 2004. Spatial scaling ectomycorrhizal fungal communities. FEMS Microbiology of microbial eukaryote diversity. Nature 432:747–750. Ecology 49:319–332. Halme P, Kotiaho JS. 2012. The importance of timing and Lindahl BD, Nilsson RH, Tedersoo L, Abarenkov K, Carlsen number of surveys in fungal biodiversity research. T, Kjøller R, Kõljalg U, Pennanen T, Rosendahl S, Stenlid Biodiversity and Conservation 21:205–219, doi: 10.1007/ J, Kauserud H. 2013. Fungal community analysis by high- s10531-011-0176-z throughput sequencing of amplified markers—a user’s Hanson CA, Fuhrman JA, Horner-Devine MC, Martiny JBH. guide. New Phytologist 199:288–299. 2012. Beyond biogeographic patterns: processes shaping Lilleskov EA, Hobbie EA, Horton TR. 2011. Conservation of the microbial landscape. Nature Reviews Microbiology 10 ectomycorrhizal fungi: exploring the linkages between (7):457–506. functional and taxonomic responses to anthropogenic N Hijmans RJ, Williams E, Vennes C. 2012. Package ‘geo- deposition. Fungal Ecology 4:174–183, doi: 10.1016/j. sphere’.Availablefrom:https://cran.r-project.org/web/ funeco.2010.09.008 packages/geosphere/. Miller SL, Larsson E, Larsson K-H, Verbeken A, Nuytinck J. Jumpponen A, Jones KL. 2009. Massively parallel 454 sequen- 2006. Perspectives in the new Russulales. Mycologia cing indicates hyperdiverse fungal communities in tempe- 98:960–970, doi: 10.3852/mycologia.98.6.960 rate Quercus macrocarpa phyllosphere. New Phytologist Nguyen NH, Smith D, Peay K, Kennedy P. 2015. Parsing 184:438–448, doi: 10.1111/j.1469-8137.2009.02990.x ecological signal from noise in next generation amplicon Kernaghan G. 2001. Ectomycorrhizal fungi at tree line in the sequencing. New Phytologist 205:1389–1393. Canadian Rockies. II. Identification of ectomycorrhizae by Oksanen JF, Blanchet FG, Kindt R, Legendre P, Minchin anatomy and PCR. Mycorrhiza 10:217–229. PR, O’HaraRB,SimposonGL,SolymosP,Stevens Kõljalg U, Dahlberg A, Taylor AFS, Larsson E, Hallenberg N, MHH, Wagner H. 2012. Package ‘vegan’.[citedxxx]. Stenlid J, Larsson K-H, Fransson PM, Kårén O, Jonsson L. Available from: http://cran.r-project.org, http://vegan.r- 2000. Diversity and abundance of resupinate thelephoroid forge.r-project.org/. fungi as ectomycorrhizal symbionts in Swedish boreal for- Parrent JL, Vilgalys R. 2007. Biomass and compositional ests. Molecular Ecology 9:1985–1996, doi: 10.1046/j.1365- responses of ectomycorrhizal fungal hyphae to elevated – 294X.2000.01105.x CO2 and nitrogen fertilization. New Phytologist 176:164 Kõljalg U, Larsson K-H, Abarenkov K, Nilsson RH, 174, doi: 10.1111/j.1469-8137.2007.02155.x Alexander IJ, Eberhardt U, Erland S, Høiland K, Kjøller Peay KG. 2014. Back to the future: natural history and the R, Larsson E, Pennanen T, Sen R, Taylor AFS, Tedersoo L, way forward in modern fungal ecology. Fungal Ecology Vrålstad T, Ursing BM. 2005. UNITE: a database provid- 12:4–9. ing Web-based methods for the molecular identification of Peay KG, Garbelotto M, Bruns TD. 2010. Evidence of dis- ectomycorrhizal fungi. New Phytologist 166:1063–1068, persal limitation in soil microorganisms: isolation reduces

Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017 doi: 10.1111/j.1469-8137.2005.01376.x species richness on mycorrhizal tree islands. Ecology Kõljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor 91:3631–3640, doi: 10.1890/09-2237.1 AFS, Bahram M, Bates ST, Bruns TD, Bengtsson-palme Peay KG, Kennedy PG, Bruns TD. 2008. Fungal community J, Callaghan TM, Douglas B, Drenkhan TM, Eberhardt ecology: a hybrid beast with a molecular master. Bioscience U, Duenas M, Grebenc T, Griffith GW, Hartmann M, 58:799–810, doi: 10.1641/B580907 Kirk P, Kohout P, Larsson E, Lindahl BD, Lucking R, Peter M, Ayer F, Cudlin P, Egli S. 2008. Belowground ecto- Martin MP, Matheny PB, Nguyen NH, Niskanen T, Oja mycorrhizal communities in three Norway spruce stands J, Peay KG, Peintner U, et al. 2013. Towards a unified with different degrees of decline in the Czech Republic. paradigm for sequence-based identification of fungi. Mycorrhiza 18:157–169, doi: 10.1007/s00572-008-0166-5 Molecular Ecology 22:5271–5277, doi: 10.1111/ Qian H, Ricklefs RE, White PS. 2005. Beta diversity of mec.12481 angiosperms in temperate floras of eastern Asia and east- Larsen MJ. 1968. Tomentelloid fungi of North America. State ern North America. Ecology Letters 8:15–22, doi: 10.1111/ University College of Forestry at Syracuse University j.1461-0248.2004.00682.x Technical Publication 93:1–157. R Development Core Team. 2013. R: a language and envir- Larsen MJ. 1974. A contribution to the taxonomy of the onment for statistical computing. Vienna, Austria: R genus Tomentella. Mycologia Memoirs 4:1–145. Foundation for Statistical Computing. MYCOLOGIA 127

Reche I, Pulido-Villena E, Morales-Baquero R, Casamayor F, Brearley FQ, Bonito G, Anslan S, Abell S, Abarenkov EO. 2005. Does ecosystem size determine aquatic bacterial K. 2014. Global diversity and geography of soil fungi. richness? Ecology 86:1715–1722. Science 346:1256688, doi: 10.1126/science.aaa1185 Rinaldi A, Comandini O, Kuyper T. 2008. Ectomycorrhizal Tedersoo L, May TW, Smith ME. 2010. Ectomycorrhizal fungal diversity: separating the wheat from the chaff. lifestyle in fungi: global diversity, distribution, and evolu- Fungal Diversity 33:1–45. tion of phylogenetic lineages. Mycorrhiza 20:217–263, doi: Smith DP, Peay KG. 2014. Sequence depth, Not PCR replica- 10.1007/s00572-009-0274-x tion, improves ecological inference from next generation Tedersoo L, Suvi T, Jairus T, Kõljalg U. 2008. Forest micro- DNA sequencing. PLoS ONE 9:e90234. site effects on community composition of ectomycorrhizal Smith ME, Douhan GW, Rizzo DM, Smith ME. 2011. fungi on seedlings of Picea abies and Betula pendula. Ectomycorrhizal community structure in a xeric Quercus Environmental Microbiology 10:1189–1201, doi: 10.1111/ woodland based on rDNA sequence analysis of sporocarps j.1462-2920.2007.01535.x and pooled roots. New Phytologist 174:847–863. Verleyen E, Vyverman W, Sterken M, Hodgson DA, De Talbot JM, Bruns TD, Smith DP, Branco S, Glassman SI, Wever A, Juggins S, Van de Vijver B, Jones VJ, Erlandson S, Vilgalys R, Peay KG. 2013. Independent Vanormelingen P, Roberts D, Flower R, Kilroy C, roles of ectomycorrhizal and saprotrophic communities Souffreau C, Sabbe K. 2009. The importance of dispersal in soil organic matter decomposition. Soil Biology and related and local factors in shaping the taxonomic struc- Biochemistry 57:282–291, doi: 10.1016/j. ture of diatom metacommunities. Oikos 118:1239–1249. soilbio.2012.10.004 Vohník M, Sadowsky JJ, Kohout P, Lhotáková Z, Nestby R, Talbot JM, Bruns TD, Taylor JW, Smith DP, Branco S, Miroslav K. 2012. Novel root-fungus symbiosis in Ericaceae: Glassman SI, Erlandson S, Vilgalys R, Liao H-L, Smith sheathed ericoid mycorrhiza formed by a hitherto undescribed ME, Peay KG. 2014. Endemism and functional convergence basidiomycete with affinities to Trechisporales. PLoS ONE 7: across the North American soil mycobiome. Proceedings of e39524, doi: 10.1371/journal.pone.0039524 the National Academy of Sciences of the United States of White TJ, Bruns T, Lee S, Taylor JW. 1990. Amplification and America 111:6341–6346, doi: 10.1073/pnas.1402584111 direct sequencing of fungal ribosomal RNA genes for phylo- Tedersoo L, Bahram M, Põlme S, Kõljalg U, Yorou NS, genetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. Wijesundera R, Ruiz LV, Vasco-palacios AM, Thu PQ, PCRprotocols:aguidetomethodsandapplications.San Suija A, Smith ME, Sharp C, Saluveer E, Saitta A, Rosas Diego, CA: Academic Press. p. 315–322. M, Riit T, Ratkowsky D, Pritsch K, Põldmaa K, Whitaker RJ, Grogan DW, Taylor JW. 2003. Geographic barriers Piepenbring M, Phosri C, Peterson M, Parts K, Pärtel isolate endemic populations of hyperthermophilic archaea. K, Otsing E, Nouhra E, Njouonkou AL, Nilsson RH, Science 301:976–978. Morgado LN, Mayor J, May TW, Majuakim L, Lodge Zak B, Larsen MJ. 1978. Characterization and classification of DJ, Lee SS, Larsson K-H, Kohout P, Hosaka K, Hiiesalu mycorrhizae of Douglas-fir. III. Pseudotsuga menziesii and I, Henkel TW, Harend H, Guo L-D, Greslebin A, Grelet Bysssoporia (Poria) terrestris vars lilacinorosea, parkssi and G, Geml J, Gates G, Dunstan W, Dunk C, Drenkhan R, sublutea. Canadian Journal of Botany-Revue Canadienne de DearnaleyJ,DeKeselA,TanDangT,ChenX,Buegger Botanique 56:1416–1424. Downloaded by [University of Wisconsin - Madison] at 20:06 13 October 2017